********************************************************************


                                                     Seminar

             Department of Systems Engineering and Engineering Management
                                  The Chinese University of Hong Kong

------------------------------------------------------------------------------------------

　

　

　

Title

:

A new method for global optimisation problems: the cross-entropy method

　

　

　

Speaker

:

Prof. Mike Maher

　

　

Institute for Transport Studies,

　

　

Leeds University

　

　

　

Date

:

November 29th, 2007 (Thursday)

　

　

　

Time

:

4:30 p.m. - 5:30 p.m.

　

　

　

Venue

:

Room 513

　

　

William M.W. Mong Engineering Building

　

　

(Engineering Building Complex Phase 2)

　

　

CUHK

　

　

　

------------------------------------------------------------------------------------------

Abstract:
　

Complex combinatorial optimisation problems arise in many areas of application of mathematical modelling. They are generally extremely difficult to solve exactly and, as a consequence, a variety of heuristic approaches have been applied to such problems including, for example, evolutionary algorithms, simulated annealing and ant colony optimisation. Each is based on a rather loose analogy with natural processes and has limited mathematical foundation. The seminar will describe an alternative, and quite recently-developed approach, known as the cross-entropy method which, it will be shown, has many attractive features including having a sound probabilistic basis, being easy to apply and requiring very little ※tuning§ of parameters. It is an iterative process that consists of generating solutions from some probability distribution whose parameter values are updated after each iteration, using information from the best solutions found in that iteration. The method will be illustrated by applying it to the challenging problem of the optimisation of signal settings on a six-arm signalised roundabout. The performance of any proposed set of timings is evaluated using a detailed macroscopic traffic model that models the formation and dissipation of queues over time and space. The results from the investigations are very encouraging, and show that the cross-entropy method has the potential to be a useful technique for tackling global optimisation problems.


-------------------------------------------------------------------------------------------

Biography:
　

Mike Maher is Professor of the Mathematical Analysis of Transport Systems at the Institute for Transport Studies at the University of Leeds, UK. His research is in the mathematical and statistical modelling of transport problems, especially in the areas of network modelling, optimisation and traffic safety modelling. Having graduated with a BA and PhD from St. John＊s College, Cambridge University, he held posts in the Institute for Transport Studies at Leeds University, and then the Department of Probability & Statistics at Sheffield University, before joining the Transport Research Laboratory in 1987. In 1994, he moved to the Transport Research Institute at Napier University, Edinburgh. He retired from Napier in early 2007 and took up his current part-time post at Leeds University. He is a Fellow of the Institute of Mathematics & its Applications, a Chartered Statistician, is the author of over 100 papers, and is on the Editorial Advisory Board for the international journal Transportation Research B.


************************* ALL ARE WELCOME ************************

　

　

　

Host

:

Prof. Leung, May-Yee, Janny

Tel

:

(852) 2609 8238

Email

:

janny@se.cuhk.edu.hk

　

　

　

Enquiries

:

Prof. Nan Chen or Prof. Sean X. Zhou

　

:

Department of Systems Engineering and Engineering Management

　

　

CUHK

Website

:

http://www.se.cuhk.edu.hk/~seg5810

Email

:

seg5810@se.cuhk.edu.hk

　

　

　

********************************************************************